U.S. patent number 7,190,854 [Application Number 11/176,711] was granted by the patent office on 2007-03-13 for methods for forming an array of mems optical elements.
This patent grant is currently assigned to Active Optical Networks, Inc.. Invention is credited to Parvinder Dhillon, Vlad J. Novotny.
United States Patent |
7,190,854 |
Novotny , et al. |
March 13, 2007 |
Methods for forming an array of MEMS optical elements
Abstract
An embodiment of the invention comprises an optical element
capable of motion in at least one degree of freedom wherein the
motion in at least one degree of freedom is enabled by serpentine
hinges configured to enable the optical element to move in at least
one degree of freedom. The embodiment further includes driving
elements configured to deflect the optical element in said at least
one degree of freedom to controllably induce deflection in the
optical element and a damping element to reduce magnitude of
resonances. Another embodiment includes a MEMS optical apparatus
comprising an optical element capable of motion in two degrees of
freedom. The two degrees of freedom are enabled by two pairs of
serpentine hinges. A first pair of serpentine hinges is configured
to enable the optical element to move in one degree of freedom and
a second pair of serpentine hinges is configured to enable the
optical element to move in a second degree of freedom. The
apparatus further includes driving elements configured to deflect
the optical element in said two degrees of freedom and a damping
element to reduce magnitude of resonances. The invention includes
method embodiments for forming arrays of MEMS optical elements
including reflector arrays.
Inventors: |
Novotny; Vlad J. (Los Gatos,
CA), Dhillon; Parvinder (Fremont, CA) |
Assignee: |
Active Optical Networks, Inc.
(Fremont, CA)
|
Family
ID: |
37833449 |
Appl.
No.: |
11/176,711 |
Filed: |
July 6, 2005 |
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
10035829 |
Oct 18, 2001 |
6963679 |
|
|
|
09981628 |
Oct 15, 2001 |
|
|
|
|
09880465 |
Sep 23, 2003 |
6625341 |
|
|
|
09865981 |
Nov 19, 2002 |
6483962 |
|
|
|
60241269 |
Oct 17, 2000 |
|
|
|
|
60211239 |
Jun 12, 2000 |
|
|
|
|
60206744 |
May 24, 2000 |
|
|
|
|
Current U.S.
Class: |
385/18 |
Current CPC
Class: |
G02B
6/3518 (20130101); G02B 6/3556 (20130101); G02B
6/357 (20130101); G02B 6/3584 (20130101) |
Current International
Class: |
G02B
6/35 (20060101) |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
Other References
US. Appl. No. 60/206,744, filed May 24, 2000, Novotny. cited by
other .
U.S. Appl. No. 60/241,269, filed Oct. 17, 2000, Novotny. cited by
other .
U.S. Appl. No. 60/211,239, filed Jun. 12, 2000, Novonty. cited by
other .
U.S. Appl. No. 09/981,628, filed Oct. 15, 2001, AO-555. cited by
other .
U.S. Appl. No. 10/035,829, filed Oct. 18, 2991, Novotny et al.
cited by other .
Walker, James A., "The Future of MEMS in Telecommunications
Netowrks." pp. R1-R7. Bell Laboratories, Lucent Technologies. Feb.
2, 2000. cited by other .
Elwenspoek, et al., "Silcion Micromachining." 1998. Cambridge
Univeristy Press. Cover page, pp. 102-111, pp. 160-161, and p. 282.
cited by other .
Maluf, Nadim, "An Introduction to Microelectromechanical Systems
Engineering." 2000. Cover page, pp. 64-73 and 186-189. cited by
other .
Rai-Choudhury, "Handbook of Microlithography, Micromachining, and
Microfabrication." 1997. SPIE Optical Engineering Press. pp.
110-116 and 120-122. cited by other .
Conant, Robert. et al., "A Flat High-Frequency Scanning
Micromirror." Solid State Sensor & Actuator Workshop, Hilton
Head, SC. 2000. cited by other .
Bishop, D.J., "Silicon Micromachines for Lightwave Networks: Can
Little Machines Make it Big?" 83 pages. Feb. 1999. cited by other
.
Bishop, D.J., "Silicon Micromachines for Lightwave Networks: The
Little Machines that will Make it Big." SPIE vol. 11, No. 2, 12
pages. Dec. 2000. cited by other .
Bishop, David J., et al., "The Rise of Optical Switching."
Scientifice American, pp. 88-94. Jan. 2001. cited by other .
Blumenthal, Daniel J., "Routing Packets with Light." Scientifice
American, pp. 96-99. Jan. 2001. cited by other .
Ehrfeld, Wolfgang, et al., "Application of Micro- and
Nanotechnologies for the Fabrication of Optical Devices." SPIE,
vol. 3276, pp. 2-14. 1998. cited by other .
Grade, John D., et al., "A Large-Defelction Electrostatic Actuator
for Optical Switching Applications." pp. 1-4. Presented at Hilton
Head 2000. cited by other .
Laor, Herzel, "Construction and Performance of a 576.times.576
Single-Stage OXC," LEOS. 3 pages. Nov. 8, 1999. cited by other
.
Laor, Herzel, "Performance of a 576.times.576 Optical Cross
Connect." NFOEC, pp. 1-5. Sep. 26, 1999. cited by other .
Lee, Sangwoo, et al., "The Surface/Bulk Micromachining (SBM)
Process: A New Method for Fabricating Released MEMS in Signle
Crystal Silicon." Journal of Mocroelectromechanical Systems, Dec.
1999. cited by other .
Milanovic, Veljko, "Optical MEMS for Optical Communications--Trends
and Developments." Adriatic Research Institute. pp. 2-6. Dec. 2001.
cited by other .
Nee, Jocelyn T., et al., "Lightwave, Optically Flat Micromirrors
for Fast Beam Steering.". Presented at IEEE/LEOS Optical MEMS
Conference in Kauai, Hawaii. 2 pages. Aug. 2000. cited by other
.
Tien, Norman C. et al., "MEMS Actuators for Silicon Micro-Optical
Elements." Proceedings of SPIE vol. 4178. pp. 256-267. 2000. cited
by other .
Bezuk, Steve, PH.D., "Flip Chip Challenges." HD Magazine. 6 pages.
Feb. 2000. cited by other .
Chen, YiJian, "Control and Shape DEsign of an Electrically-Damped
Comb Drive for Digital Switches." MOEMS and Miniaturized Systems.
Proceedings of SPIE, vol. 4178. pp. 387. cited by other .
Dawson, J.M. et al., "MEMS Feedbach Control Using Through-Wafer
Optical Device Monitoring." Proc. of SPIE vol. 4178, MOEMS and
Miniaturized Systems. Sep. 2000. cited by other.
|
Primary Examiner: Bovernick; Rodney
Assistant Examiner: Stahl; M. J.
Attorney, Agent or Firm: Silicon Edge Law Group, LLP Behiel;
Arthur J.
Parent Case Text
CROSS-REFERENCE TO RELATED APPLICATIONS
This U.S. patent is a continuation of application Ser. No.
10/035,829 entitled "Micro-Opto-Electro-Mechanical Switching
System," by Vlad Novotny and Parvinder Dhillon, filed Oct. 18, 2001
now U.S. Pat. No. 6,963,679, application Ser. No. 10/035,829 is: 1.
a continuation of application Ser. No. 09/981,628, filed Oct. 15,
2001 now abandoned, also by Vlad Novotny and Parvinder Dhillon; 2.
a continuation-in-part of application Ser. No. 09/865,981, now U.S.
Pat. No. 6,483,962 B2, i ssued on Nov. 19,2002, entitled "Optical
Cross Connect Switching Array System with Optical Feedback" filed
on May 24, 2001 and invented by Vlad J. Novotny, which claims
priority under 35 U.S.C. .sctn.119(e) from U.S. patent application
Ser. No. 60/206,744, entitled, "Optical Cross Connect Switching
Array Systems With Optical Feedback Control" filed May 24, 2000;
and from U.S. patent application Ser. No. 60/241,269, entitled,
"Micro-Electro-Mechanical Systems for Optical Switches and
Wavelength Routers," by Vlad J. Novotny and Parvinder Dhillon,
filed Oct. 17, 2000. 3. a continuation-in-part of application Ser.
No. 09/880,456, now U.S. Pat. No. 6,625,341 B1, issued Sep. 23,
2003, entitled: "Optical Cross Connect Switching Array System with
Electrical and Optical Position Sensitive Detection", invented by
Vlad J. Novotny, filed Jun. 12, 2001. U.S. Pat. No. 6,625,341 B1 is
a continuation-in-part of aforementioned application Ser. No.
09/865,981, filed May 24, 2001, U.S. Pat. No. 6,483,962 B2; and
claims priority under 35 U.S.C. .sctn.119(e) from U.S. patent
application Ser. No. 60/211,239, entitled "Optical Cross Connect
Switching Array Systems With Multiple Optical And Electrical
Position Signal Detectors," by Vlad J. Novotny, filed Jun. 12,
2000. Each of the above-referenced patents and applications is
hereby incorporated by reference.
Claims
What is claimed is:
1. A method for forming an array of MEMS optical elements, the
method comprising: providing a single crystal silicon on insulator
(SOI) wafer having a layered structure comprising a silicon wafer
layer having an internal oxide layer formed thereon and having a
device silicon layer formed on the internal oxide layer, wherein a
top surface of the device silicon layer has formed thereon a top
oxide layer, and wherein the bottom surface of the silicon wafer
layer has formed thereon a bottom oxide layer; forming a bottom
photoresist layer on the bottom oxide film layer having openings
defining a bottom pocket; forming a top photoresist layer on the
top oxide film layer having openings defining a hinge region and
open regions; removing the top oxide layer in the hinge and open
regions defined by the openings in the top photoresist layer to
expose the hinge region of the device silicon layer; forming a
second photoresist layer on a top surface of the SOI wafer, the
second photoresist layer patterning the hinge region of the device
silicon layer so that a hinge can be formed; etching the patterned
hinge region to remove portions of the device silicon layer forming
recessed portions defining the hinge; removing the second
photoresist layer, thereby exposing the underlying top oxide layer
as a hard mask layer having openings in the hinge and open regions;
etching the device silicon layer through the openings in the hard
mask wherein the recessed portions are etched until the internal
oxide layer is reached, and wherein the unetched surfaces are
partially etched leaving a portion of the unetched surfaces in
place to define a thickness of the hinge; etching the bottom
surface of the SOI wafer through openings in the bottom oxide layer
to remove material from the silicon wafer layer to form a pocket
region defining a movable optical element supported by the hinge;
and etching the SOI wafer to remove the internal oxide layer in the
pocket region.
2. The method of claim 1, further comprising etching material from
a separation line region between adjacent optical elements to
enable the array to be separated into a plurality of smaller
arrays.
3. The method of claim 2, wherein etching material from a
separation line region comprises the operations of dry etching to
remove material from the separation line region; laser cutting in
the separation line region; and cleaving in the separation line
region into separate arrays of desired sizes.
4. The method of claim 2, wherein etching material from a
separation line region comprises the operations of laser cutting in
the separation line region and cleaving in the separation line
region into separate arrays of desired sizes.
5. The method of claim 4 wherein etching material from a separation
line region comprises the operations of dry etching to remove
material from the separation line region and cleaving in the
separation line region into separate arrays of desired sizes.
6. The method of claim 4 further comprising assembling the
separated arrays with another wafer having formed thereon
appropriate driving elements and control circuitry, and packaging
the assembled arrays.
7. The method of claim 6 wherein packaging the assembled arrays
comprises hermetically packaging the assembled arrays.
8. A method for forming an array of MEMS optical elements, the
method comprising: providing a wafer having top surface and a
bottom surface, the top surface having formed thereon a top
insulating layer and the bottom surface having formed thereon a
bottom insulating layer; forming a bottom mask layer on the bottom
oxide film layer having openings defining a bottom pocket; forming
a top mask layer on the top insulating layer having openings
defining hinge regions and open structures; first etching to remove
the top insulating layer in hinge and open regions defined by the
openings in the top mask layer exposing a hinge region; forming a
second mask layer on the top surface, the second mask layer
patterning the hinge region so that a hinge can be formed; second
etching the patterned hinge region and open region to form a hinge
in the wafer; removing the second mask layer, thereby exposing the
underlying top insulating layer as a hard mask layer having
openings in the hinge and open regions; third etching the wafer
through the openings in the hard mask wherein the recessed portions
are etched in a timed etch leaving a portion of the unetched
surfaces in place as hinges, thereby defining hinge thickness;
fourth etching the bottom surface of the wafer through openings in
the bottom oxide layer to remove material from the silicon wafer to
form a pocket region defining a movable optical element supported
by hinges; and forming a reflective layer on at least one surface
of the movable optical element.
9. The method of claim 8 further including a fifth etching to
remove material from a separation line region thereby enabling the
structure to be separated into arrays of a desired size.
10. The method of claim 9 wherein the operation of fifth etching
comprises the operations of dry etching to remove material from the
separation line region; laser cutting in the separation line
region; and cleaving in the separation line region into separate
arrays of desired sizes.
11. The method of claim 9 wherein the operation of fifth etching
comprises the operations of laser cutting in the separation line
region; and cleaving in the separation line region into separate
arrays of desired sizes.
12. The method of claim 9 wherein the operation of fifth etching
comprises the operations of dry etching to remove material from the
separation line region; and cleaving in the separation line region
into separate arrays of desired sizes.
13. The method of claim 9 further comprising assembling the
separated arrays with another wafer having formed thereon
appropriate driving elements and control circuitry, and packaging
the assembled arrays.
Description
FIELD OF THE INVENTION
The invention generally described herein relates to the design and
fabrication of micro-optic devices. In particular, the present
invention pertains to micro-electro-mechanical systems (MEMS)
optical assemblies used in fiber optic switching arrays, wavelength
routers, laser scanners, bar code scanners, variable optical
attenuators (VOA), wavelength tunable lasers, and other related
devices. More particularly, the present invention pertains to the
design, structure and fabrication of MEMS reflectors and hinges
used in fiber optic switching devices.
BACKGROUND
As is well known, fiber optic technology is a rapidly growing field
with vastly expanding commercial applicability. As with all
technologies, fiber optic technology is faced with certain
practical difficulties. In particular, the design and fabrication
of arrays of optical elements that enable the efficient switching
and coupling between input optical elements and output optical
elements in an optical network is a significant consideration of
designers, manufacturers, and users of optical systems. Optical
systems commonly use laser generated light beams, to carry
information through optical fibers and are directed through complex
optical paths with the assistance of optical switching elements,
routers and other like components. Other applications include
wavelength routers that demultiplex incoming signals into
individual wavelength and then switch in the nonblocking fashion
single wavelengths between outputs, laser beam deflectors in laser
printers, bar code reading devices and others.
FIG. 1 schematically illustrates a portion of a fiber optic network
100. In the depicted embodiment, network 100 routes optical signals
through fiber optic lines L from node to node to form an interlaced
ring-mesh network structure. Many other configurations of network
structures are possible. In the depicted embodiment, the fiber
optic lines L are interconnected at optical nodes 101, 102, 103,
104, 105, and 106. The optical signals are directed to their
desired destination by optical switching. Typically, this switching
is accomplished at the optical nodes 101, 102, 103, 104, 105, and
106 (also referred to herein as switching nodes). Each switching
node 101, 102, 103, 104, 105, and 106 accommodates a plurality of
fiber optic lines L which comprise input fibers and output fibers.
It is the selecting of and switching between these input fibers and
output fibers that define the optical paths which route optical
signals to their desired target destinations.
FIG. 2(a) is a simplified schematic illustration showing an
overview of bi directional optical cross-connect switching array
system 200. The system 200 includes fiber arrays 202 and 204 for
passing light beams into and out of the switching array system 200.
Each fiber array 202, 204 comprises a plurality of fiber optic
transmission lines (a portion of which are shown here by fibers
210, 211, 220, and 221). For convenience, fiber array 202 shall be
referred to as an incoming fiber array 202 and the fiber array 204
shall be referred to as an outgoing fiber array 204. However, it
should be remembered that due to the bi directional nature of the
switching array system 200, the terms incoming and outgoing are
relative.
Light beams carry information throughout the optical network. The
light beams are directed to their final destination by passing
through switching array systems 200 which direct the light beams to
the desired destination. Electronic control circuitry 230 is used
to dynamically control the switch 200 configuration. The control
circuitry 230 can include, among other elements, position sensitive
detectors, demultiplexing circuitry, photodetectors, position
sensing detectors, amplifiers, decoding circuitry, servo
electronics, digital signal processors, communication hardware, and
an application programming interface. The control circuitry directs
entering light beams to the desired exit fibers.
The following simplified illustration describes how a light beam
can be switched from one of the incoming fibers in array 202 to a
selected one of the fibers in array 204. Such description is also
applicable to switching a light beam between any selected fiber in
array 204 to a selected fiber in array 202.
In the depicted illustration, the light beam 231 exits the fiber
210 (and in preferred embodiments, passes through a lens array (not
shown) so that the beam propagates without significant divergence)
onto the reflector array 218. Servo electronics of the control
circuitry 230 initiate deflection in a reflector 218' of the
reflector array 218 to direct the light beam 231 along an optical
path 232 to a desired fiber 220 (in fiber array 204) using a signal
from position detection array 234. By changing the deflection of
the reflectors (e.g., 218') of the reflector array the light beams
can be switched to enter any selected outgoing fiber 204. Also, the
deflection of each of the reflectors 218' can be altered in very
small ways to fine tune light beam optical characteristics. The
reflector 218' deflection can be adjusted in response to
instructions contained within the data streams of the light beam
231. Alternatively, reflector 218' deflection can be adjusted in
response to instructions provided externally via an application
programming interface of, for example, the control circuitry. Other
methods of adjusting reflector 218' deflection known to those
having ordinary skill in the art can also be used.
A light beam can be switched from one outgoing fiber to another
outgoing fiber, by changing reflector deflection angle. For
example, if light beam 231, 232 is to be switched from fiber 220
into another outgoing fiber 221, the controller circuitry 230 sends
appropriate instructions to the servo electronics which reposition
the reflector 218' so that beam 231 is redirected along optical
path 233 to fiber 221. Typically, the beams (e.g., 232, 233) pass
through a lens array (not shown) which focuses and couples the
light beam (here 233) into the outgoing fiber (here 221). It should
be noted that although fibers have heretofore been referred to as
belonging to the incoming fiber arrays 202 or the outgoing fiber
arrays 204, such fiber arrays are bi-directional. In such
bi-directional embodiments, light beams also travel from the
outgoing fibers in the outgoing fiber array 204 to incoming fibers
in the incoming fiber array 202. This is done in the same way as
light beams traveling from incoming fibers in the incoming fiber
array 202 to outgoing fibers in the outgoing fiber array 204. Also
shown in FIG. 2(a) are the position-sensitive-detectors 234, which
feed the position-error-signals to the controller circuitry
230.
The switching array system 200 is shown as one-dimensional in the
embodiment of FIG. 2(a) for clarity. In preferred embodiments the
aforementioned arrays are two dimensional. For example, in an
embodiment with a two-dimensional reflector array 218, there are
rows and columns, or some other two-dimensional arrangement of
reflectors. The other arrays and alignment structures are similarly
two-dimensional in some embodiments. In addition, the overall
system is shown as two-dimensional in FIG. 2(a). In preferred
embodiments, the system is three-dimensional, as the additional
dimension in and out of the plane of the paper can be
advantageously used to position the various components and minimize
the dimensions of the hardware.
It should be noted that although FIG. 2(a) depicts the switching
device 200 as having a single reflector array 218, many embodiments
include two or more reflector arrays instead of just one with or
without additional plane reflectors. One such embodiment is
schematically illustrated in FIG. 2(b). FIG. 2(b) is a simplified
schematic illustration showing an overview of two-reflector array
bi-directional optical cross-connect switching array system 201.
The system 201 includes fiber arrays 202 and 204 for passing light
beams into and out of the switching array system 201. The fiber
arrays 202, 204 include a plurality of fiber optic transmission
lines (a portion of which are shown here by fibers 210, 211, 220,
and 221). Here, the incoming light beam 234 is directed toward a
first reflector array 217 which reflects the beam 234, 235 onto a
second reflector array 219 and then into the desired outgoing fiber
(here, 221). Switching may be accomplished by altering the
deflection of the reflectors of the first reflector array 217 or by
altering the deflection of the reflectors of the second reflector
array 219 or by altering the deflection of the reflectors of the
first reflector array 217 and the reflectors of the second
reflector array 219 at the same time. In this example, the path of
light beam 234 is altered by the deflection of first reflector 217'
which directs the light beam 234 onto the altered beam path 236
onto second reflector 219' and into outgoing fiber 220.
Additionally the control circuitry (not shown) controls the
reflectors of both the first reflector array 217 and the second
reflector array 219. Although structurally somewhat different from
the previously discussed embodiment 200, the principles of
operation of such multiple reflector array switches 201 are
similar. Similar switching functions can be performed using
alternative switching configurations. For example, one embodiment
can use combined first and second sets of movable reflectors and
one fixed reflector. An optical beam can be switched by reflection
of an input beam from a movable reflector onto a fixed reflector
and from this reflector back onto a movable reflector and into
output fiber. Number of reflectors in the combined array is the
same as total number of reflectors in two physically separate
arrays. Many other configurations are used and known by those
having ordinary skill in the art.
MEMS switching arrays can also be used in wavelength routers. One
embodiment of such a wavelength router is depicted in FIG. 3. Using
wavelength division multiplexing light beams of several wavelengths
can be optically transmitted using the same fiber. For example, a
single fiber can carry light beams comprising k signals at k
wavelengths. These light beams of many wavelengths are coupled from
a fiber 331 into a wavelength division demultiplexer 334. The
demultiplexer 334 can be based on arrayed waveguide gratings,
interference filters, or fiber Bragg gratings. The illustration of
FIG. 3 uses an arrayed waveguide grating 334 as a wavelength
division demultiplexer. Multi wavelength light beam 345 enters into
the first free space region 335, is separated into individual
wavelengths in grating 333 and exits through the second free space
region 336 where light beams at k wavelengths are spatially
separated. Light at each specific wavelength is coupled into linear
fiber array that directs light beams onto a lens array 342.
Relatively collimated light beams such as 343 and 344 propagate
toward mirrors of the first array 337. The light at each specific
wavelength is reflected from one mirror in the first array 337 onto
a specific mirror of the second mirror array 338 from which the
light is directed onto focusing lenses 339 and into a selected
output fiber 351. The mirror arrays 337 and 338 can be
one-dimensional arrays in order to match the spatial distribution
of the light beams or two-dimensional arrays. Mirror arrays 337 and
338 are formed by bi-axial (bi-axially actuated) mirrors.
SUMMARY
In accordance with the principles of the present invention, one
embodiment of the invention comprises an optical element capable of
motion in at least one degree of freedom wherein the motion in at
least one degree of freedom is enabled by serpentine hinges
configured to enable the optical element to move in the at least
one degree of freedom. The embodiment further includes driving
elements configured to deflect the optical element in said at least
one degree of freedom to controllably induce deflection in the
optical element and a damping element to reduce magnitude of
resonances
Another embodiment includes a MEMS optical apparatus comprising an
optical element capable of motion in two degrees of freedom. These
degrees of freedom are enabled by a first pair of serpentine hinges
that is configured to enable the optical element to move in one
degree of freedom and a second pair of serpentine hinges that is
configured to enable the optical element to move in a second degree
of freedom. The apparatus further includes driving elements
configured to deflect the optical element in said two degrees of
freedom and a damping element to reduce magnitude of
resonances.
Another embodiment includes a MEMS optical apparatus comprising in
combination a support structure, a movable optical element, at
least one pair of serpentine hinges, driving elements positioned
such that activation of the driving elements can controllably
induce deflection in the movable optical element and a damping
element. The combination comprising means for inducing a damped
rotation of the movable optical element about an axis of rotation
defined by each of the at least one pair of serpentine hinges.
A method embodiment for forming an array of MEMS optical elements
comprises: providing a silicon-on-insulator (SOI) wafer.
Photoresist masking the top and bottom surfaces with appropriate
patterning. First etching to remove the top oxide layer in hinge
regions defined by the openings in the top photoresist layer
exposing a hinge region of the device silicon layer. Forming a
second photoresist layer patterning the hinge region of the device
silicon layer so that a hinge can be formed. Second etching the
patterned hinge region to remove portions of the device silicon
layer forming recessed portions and such that unetched surfaces
correspond to a hinge. Removing the second photoresist layer,
thereby exposing the underlying top oxide layer as a hard mask
layer having openings in the hinge region. Third etching the device
silicon layer through the openings in the hard mask wherein the
recessed portions are etched until the internal oxide layer is
reached wherein the previously unetched surfaces are partially
etched leaving a portion of the unetched surfaces in place as
hinges. Fourth etching the bottom surface of the SOT wafer to form
a pocket region and a separation line region. Fifth etching the SOT
wafer to remove the internal oxide layer in the pocket region.
Forming a reflective layer on at least one surface of the movable
optical element, and a sixth etching to remove material from the
separation line region to complete the separation line thereby
enabling the substrate to be separated into arrays of a desired
size.
These and other aspects and advantages of the invention will become
apparent from the following detailed description and accompanying
drawings which illustrate, by way of example, the principles of the
invention.
BRIEF DESCRIPTION OF THE DRAWINGS
For a fuller understanding of the invention, reference is made to
the accompanying drawings in the following Detailed Description. In
the drawings:
FIG. 1 is a figurative illustration of an optical network.
FIGS. 2(a) and 2(b) are simplified schematic illustrations of a
single reflector array and two reflector array optical switch
embodiments.
FIG. 3 is a simplified schematic illustration of an embodiment of a
wavelength router.
FIG. 4(a) is a top down view of an embodiment of a reflector
array.
FIGS. 4(b) and 4(c) are top down views of an embodiment of a
reflector assembly.
FIGS. 4(d) and 4(e) are cross section views of a portion of the
embodiment shown in FIGS. 4(b) and 4(c).
FIGS. 5(a) 5(e) are top down views of serpentine hinge embodiments
in accordance with the principles of the present invention.
FIG. 5(f) is perspective view of a serpentine hinge embodiment in
accordance with the principles of the present invention.
FIGS. 5(g), 5(h) and 5(i) are plan and cross-sectional views of the
hinge embodiments having damping material applied in accordance
with the principles of the present invention.
FIGS. 6(a) and 6(b) are plan views of a reflector assembly
embodiment in accordance with the principles of the present
invention particularly depicting frame, mirror, and serpentine
hinge elements and the associated underlying driving elements.
FIGS. 7(a) and 7(b) are plan views of another reflector assembly
embodiment in accordance with the principles of the present
invention particularly depicting frame, mirror, and radial
serpentine hinge elements and the associated underlying driving
elements.
FIG. 7(c) is a plan view of a radial serpentine hinge in accordance
with the principles of the present invention.
FIGS. 8(a) and 8(b) are plan views of another reflector assembly
embodiment in accordance with the principles of the present
invention particularly depicting frame, mirror, and
circumferentially curved serpentine hinge elements and the
associated underlying driving elements.
FIG. 9(a) is a plan view of a variable spring constant serpentine
hinge embodiment in accordance with the principles of the present
invention FIGS. 9(b) and 9(c) are plan views of another reflector
assembly embodiment in accordance with the principles of the
present invention particularly depicting frame, mirror, and
circumferentially curved variable spring constant serpentine hinge
elements and the associated underlying driving elements.
FIGS. 10(a) and 10(b) are plan views of another reflector assembly
embodiment in accordance with the principles of the present
invention particularly depicting multiple frames, mirror, straight
hinge elements, serpentine hinge elements, and the associated
underlying driving elements.
FIGS. 11(a) 11(m) depict a series of cross-section views of a
substrate upon which a reflector embodiment is being formed in
accordance with the principles of the present invention, each
Figure illustrating various steps of a fabrication process.
Reference numerals refer to the same or equivalent parts of the
present invention throughout the several figures of the
drawings.
DETAILED DESCRIPTION
The present invention has been particularly shown and described
with respect to certain preferred embodiments and specific features
thereof. The embodiments set forth herein below are to be taken as
illustrative rather than limiting. It should be readily apparent to
those of ordinary skill in the art that various changes and
modifications in form and detail may be made without departing from
the spirit and scope of the invention.
FIG. 4(a) figuratively depicts a plan view of an embodiment of an
M.times.N array 400 of movable optical elements 401 (where M and N
represent integer values from 1 to m and from 1 to n,
respectively). Where the movable optical elements 401 are
reflectors, such an array 400 can be incorporated into an optical
switching device in accordance with the principles of the present
invention. The array contains a plurality of movable optical
elements 401 formed on the substrate or support structure of the
array 400. These optical elements can comprise a wide range of
optical components including, but not limited to reflectors
(mirrors), blocking optics (which block the transmission of light),
filters, gratings and lenses. Such movable optical elements serve a
number of purposes and can be incorporated into numerous optical
devices including optical switches. Such movable optical elements
401 can be movable about one axis or about two axes (so-called
bi-axial optical elements). Throughout this patent these movable
optical elements 401 will be discussed in the context of
reflectors. It should be appreciated by those having ordinary skill
in the art that the movable optical elements 401 described herein
as reflectors can be interchanged with other optical elements,
including but not limited to any of the aforementioned optical
elements. Thus, the optical element array of FIG. 4(a) will be
described as reflector array 400. Each reflector array 400 includes
M.times.N reflector assemblies 401 formed on the substrate
structure of the reflector array 400. The inventors contemplate
many uses for such reflector arrays including, but not limited to
single reflector array switching devices and two reflector array
switching devices, as well as wavelength routers incorporating
single or double reflector arrays.
FIG. 4(b) schematically illustrates aspects of a bi-axial reflector
assembly 401 capable of deflection in two degrees of freedom.
Embodiments for rotating in one degree of freedom are also
contemplated by the inventors. The depicted reflector assembly 401
includes a reflective element 407, commonly referred to as a
mirror. The mirror 407 is supported in a frame 408 by a pair of
mirror hinges 404. The hinges depicted here are schematic in
nature. The preferred hinge embodiments are discussed in greater
detail below. The pair of mirror hinges 404 supports the mirror 407
such that an axis of rotation (here, for example, rotation about an
X-axis) is defined. The frame 408 is supported in the substrate
structure 405 of the reflector array 400 by another pair of frame
hinges 406. Typically, the mirror 407 is positioned inside a recess
in the substrate structure 405 of the reflector array 400 such that
the mirror 407 has clearance to be tilted. Alternative embodiments
for the mirror 407 position the mirror 407 so that it is raised
above the surface of the substrate structure 405 of the reflector
array 400.
The pair of mirror hinges 406 supports the frame 408 such that
another axis of rotation (here, for example, rotation about a
Y-axis) is defined. Typically, the pairs of hinges 404, 406 define
substantially perpendicular axes of rotation. Thus,
three-dimensional motion can be achieved in the reflector
assemblies 401. Simpler, reflector assemblies can also be
constructed. Such assemblies only rotate about a single axis. These
reflector assemblies find utility in many applications including
smaller optical switches and in so-called digital (on-off)
switching arrays. Such arrays only require rotation about a single
axis. Generally speaking, such reflector assemblies 401 are driven
by electrostatic, electromagnetic, piezoelectric or thermal driving
elements. Electrostatic actuators are commonly fabricated
underneath the mirror 407 and frame elements 408. These driving
elements are typically controlled by the control circuitry of the
switch. The control circuitry that drives the driving elements can
be formed directly underneath the driving elements as part of the
fabrication process or elsewhere on the reflector array 400.
Alternatively, the control circuitry that drives the driving
elements can be formed completely separate from the array 407 and
connected later.
FIG. 4(c) schematically illustrates driving elements (or drive
elements) 407', 407'', 408', and 408''. In the depicted
illustration, the drive elements are, for example, positioned
beneath the moving parts. The driving elements 407', 407'' rotate
the mirror 407 about the X-axis, and driving elements 408', and
408'' rotate the frame 408 about the Y-axis. The driving elements
407', 407'', 408', and 408'' are typically constructed of parallel
plate capacitors. Driving control electronics can be included below
driving elements 407', 407'', 408', and 408'' for larger reflector
arrays and connected to the driving elements using, for example,
vias. In other embodiments, the driving electronics can be on
separate wafers and leads can be routed on the surface to the
driving elements 407', 407'', 408', and 408''.
FIG. 4(d) is a cross section view of a portion of the embodiment
shown in FIG. 4(c). The driving elements 407', 407'', 408', 408''
depicted in this embodiment is preferably constructed having a
slightly smaller size than the frame 408 and reflector 407
elements. During operation, the frames and mirror elements have
been known to rotate too much causing the outer edges of the frames
and mirror elements to make mechanical contact with the underlying
drive elements. This electrically short circuits the system and can
permanently damage the system components. Thus, if the driving
elements have a slightly smaller dimension, especially along the
outer edges, excessive deflection of the frames and mirror elements
will not result in shorting of the system. As depicted in FIG.
4(c), the drive elements are smaller (especially at the distal
edges) than the movable elements, as indicated by the heavy
boundary lines. Therefore, even if the movable elements contact the
underlying structure there will be no electrical contact between
the drive elements 407', 407'' and the depicted reflector 407. The
same can be said of drive elements 408', 408'', and frame 408. Such
sizing of the drive elements can be utilized with respect to all
the embodiments set forth herein. Implementation in FIG. 4(d) has
the gap 415 between mirror element 407 and frame 408 and driving
elements 407', 407'', 408' and 408'' defined by the thickness of
the wafer 412 minus thickness of mirror 407. When smaller gaps 416
shown in FIG. 4(e) are required, the bottom chip 414 with driving
elements is etched with trenches 417. Metal coatings 410 and 411
are also shown in FIGS. 4(d) and 4(e). The frame 408 does not
require metal coating as the silicon material is extrinsic, having
dopants in silicon for all these devices. When wafer 414 is
electrically conducting, dielectric film is deposited on its
surface in areas where the physical contact occurs between 412 and
414 in order to electrically insulate parts 412 and 414.
Although depicted here as parallel plate electrostatic actuators,
driving elements in accordance with the principles of the present
invention may be of many different types of actuators known to
those having ordinary skill in the art can be used. For example,
other types of actuators such as electrostatic rotational comb
actuators, electromagnetic actuators, piezoelectric actuators or
thermally driven actuators can be used. Although the depicted
embodiment shows the mirror 407 as circular, the mirror 407 can
have any shape.
Hinge design is an important aspect of the high performance
reflector assemblies. The length, width, thickness, and cross
sectional shape of hinges determine the stiffness and consequently
the driving signals (voltages in case of electrostatic actuators)
required to achieve desired deflections in the reflectors and the
desired frequency response of the actuator. The torsional hinge
stiffness is proportional to hinge thickness, to the third power of
hinge width and inversely proportional to hinge length. The bending
hinge stiffness is proportional to the third power of hinge
thickness, to hinge width and inversely proportional to hinge
length. The hinge stiffness has to be low enough to provide
sensitive deflections but also high enough to exhibit high
frequency resonances. The hinge must also be robust enough to be
manufactured with a high yield and withstand the conditions of a
normal operating environment. Additionally, if the reflector was
constructed so that both the mirror and the hinges have the same
low thickness, the lack of flatness of the mirror would lead to
excessive wavefront distortions in light reflected by the mirror.
Consequently, in most cases, the mirror thickness will be greater
than hinge thickness. Therefore, fabrication processes should be
capable of generating these two different thicknesses. In addition,
hinge width is limited by processing (lithography and etching) and
reasonable widths do not lead to acceptably low stiffness, unless
the length of hinges is much greater than that which straight
hinges can provide.
The principles of the present invention address this problem by
using a serpentine hinge structure. Serpentine hinges can include
one, two, three, four or more "windings". The inventors contemplate
that n windings can be used in the hinges, where n is equal to or
greater than one. FIGS. 5(a) 5(d) depict examples of such single
(5(a)), double (5(b)), triple (5(c)), and quadruple (5(d))
serpentine hinges. FIG. 5(e) shows a single winding hinge 520. Each
winding includes two arms (shown here inside the dashed line boxes
521, 522). In embodiments having many windings, the arms snake
continuously from one arm to another for the entire length of the
hinge. The windings include a pair of shafts 523, 524 which connect
the hinges 520 to the larger array elements. The arms 521, 522
extend in a direction transverse to that of the axis of rotation
for the hinge. FIG. 5(f) is a perspective view of a portion of a
double serpentine hinge 500. The thickness 501 of the hinge
typically ranges from about 3 (micron) to about 50 km. Thicker
embodiments can be fabricated and, for some embodiments, are
favored. However, the most preferred thickness is in the range of
about 5 10 This is in comparison to a typical mirror embodiment
which is in the range of about 10 50 thick. The width 502 of the
hinge typically ranges from about 2 to about 10 Again, embodiments
having greater widths can be fabricated. However, the preferred
width is in the range of about 5 10 The length (defined here by the
dashed line) 503 of the serpentine hinge 500 can be any length.
Embodiments having lengths in the 100's or even 1000's of microns
being preferred. Of course, the length 503 of the serpentine hinge
500 depends on the number of windings in the hinge.
Damping is an advantageous feature that can add to the utility of
each of the embodiments disclosed herein. One example of such a
means is a thin coating of a damping agent applied onto the hinges.
Such damping agents when dried (or cured) act as a damping factor
which reduces resonances in the optical structures disclosed
herein. Such damping agents are typically polymeric materials.
Suitable materials include, but are not limited to silicones and
elastomer materials for example, di-methylsilicone, polyurethane,
polyisobutene-co-isoprene, and polybutadiene-co-acrylonitrile. Such
damping agents are coated onto the hinges and cured. Alternatively,
the damping agents are dried until the volatile constituents
outgas. Typically, such damping agents are applied onto the hinges
and part of the adjoining support structures. Such damping agents
can be applied using, for example, an ink jet dispensing in any
desired pattern and quantities over the hinge surfaces. Curing can
be with room temperature, elevated temperature or exposure to
ultraviolet radiation, electron beams, or a combination of these
methods. In some embodiments, the damping agent is applied to the
hinge in smaller quantities, forming isolated "islands" of damping
material on the surface of the hinges. The amount of material
applied to the hinges can depend on many factors, including,
material type, amount of adjustment necessary, thickness of
material, method of application, and other factors.
FIG. 5(g) is a drawing showing a layer of the viscoelastic material
551 applied over the hinge 520 between the frame 552 and the
adjoining support structure 553. This material is applied to fine
tune the device performance after its fabrication by providing
means of adjusting the damping while monitoring the device
characteristics. The viscoelastic sheet has an adhesive coating on
one side and the appropriately sized pieces are applied over the
hinge area.
FIG. 5(h) shows a variation on the use of the viscoelastic material
561 so as to cover only the area of the hinge 520. Application
method is based on ink jet dispensing in any desired pattern and
quantities over the hinge surfaces. Curing can be with room
temperature, elevated temperature or exposure to ultraviolet
radiation, or a combination of these methods.
FIG. 5(i) shows yet another variation on the extent of coverage of
the viscoelastic material 571 over the hinge 520. In this case the
material is applied to the hinge in smaller quantities, forming
isolated islands 571. The amount of material applied to the hinge
will depend on many factors, including, material type, amount of
adjustment necessary, thickness of material, method of application,
and other factors. Some examples of the elastomer materials are
Di-methylsilicone, polyurethane, polyisobutene-co-isoprene, and
polybutadiene-co-acrylonitrile.
FIG. 6(a) is a plan view illustrating one embodiment of a reflector
assembly 600 in accordance with the principles of the present
invention. The mirror 607 is held in the frame 608 by a pair of
serpentine mirror hinges 609 (or mirror hinges). The serpentine
mirror hinges 609 are depicted as having two windings. Other
embodiments can include 1 to n windings. The frame 608 is suspended
in the array substrate 603 by a second set of serpentine hinges 606
(also referred to as frame hinges). As with the mirror hinges 609,
the frame hinges 606 can comprise from 1 to n windings
FIG. 6(b) shows a drive assembly 601 which lies just underneath the
mirror/frame/hinge structure depicted in FIG. 6(a). FIG. 6(b)
depicts the drive elements 607', 607'', 608', and 608'' that are
the components of electrostatic actuators. The drive elements 607',
607'' interact with the mirror 607 to provide positive and negative
deflection about a axis. The drive elements 608', 608'' interact
with the frame 608 to provide positive and negative deflection
about an X-axis. The drive elements 607', 607'', 608', and 608''
are shaped such that they do not interfere with the hinges 609 and
606. This typically means that the drive elements 607', 607'',
608', and 608'' are not formed under the hinges. Also, it is
preferred that the drive elements 607', 607'', 608', and 608'' be
sized such that, in the event of excessive deflection of the
movable mirror and frame components, no contact is made between the
drive elements and the movable components. This is typically
avoided by reducing the size of the drive elements such that the
outer edges of the movable components will not contact the drive
elements even in the event of excessive deflection. Thus, the drive
elements 607', 607'', 608', and 608'' are slightly smaller, in the
regions 654, 655, 656, 657, respectively, than the overlying mirror
607 and frame 608. Such precautions can be implemented into each
embodiment discussed herein.
Another advantageously constructed embodiment is depicted in FIG.
7(a). This embodiment meets the challenge of packing the mirror
707, two sets of hinges 704 and 706, and frame 708 into as small an
area as possible, so that optical components can be smaller and the
overall dimensions of the cross connect switching system can be
reduced. FIG. 7(b) shows a layer of the reflector assembly 701,
which is formed just underneath the mirror/frame/hinge structure
depicted in FIG. 7(a). FIG. 7(b) depicts the drive elements 707',
707'', 708', and 708''. As discussed above with respect to the
embodiment of FIG. 6, the drive elements 707', 707'' interact with
the mirror 707 to provide deflection about a first axis. And the
drive elements 708', 708'' interact with the frame 708 to provide
deflection about a second axis. The fabrication of such structures
will be discussed in some detail hereinbelow.
With continued reference to FIG. 7(a), by packing more optical
components on a given reflector array, smaller arrays may be
constructed. Smaller reflector arrays allow a larger number of
devices to be built on a given wafer, thus reducing the cost of
these reflector arrays and the deflection angles required for
switching. Furthermore, smaller structures have higher resonance
frequencies, which improve switching and addressing times for the
reflector array. Also, smaller reflector arrays enable shorter
optical paths within switching devices. Due to the shorter optical
paths possible with such embodiments, lower resolution position
sensing systems can be used, thereby reducing cost. The depicted
pairs of serpentine hinges 704, 706 each have two windings. In
order to achieve more compact serpentine hinges, portions of the
windings are folded into a rectangular conformation, with the arms
of each winding being fabricated to include proximal folds that are
oriented such that they are parallel to the axis of rotation. An
embodiment of such a radial serpentine hinge 704 is depicted in
FIG. 7(c). The hinge 704 permits rotation (shown by the arrow) of
the mirror 707 about axis of rotation. In the previous embodiment,
the arms of each winding extend in a direction transverse to the
axis of rotation. In the depicted embodiment, the arms 710, 711,
712, 713 of each winding are formed such that a portion of the arms
(also referred to as the folded portion) extends approximately
parallel to the axis of rotation. In the depicted embodiment, the
inner folded arms (e.g., 712 and 713) are shorter than the outer
folded arms (e.g., 710 and 711). In other embodiments having more
windings, the arms are progressively longer and longer, the further
the folded arms are from the axis of rotation x. One objective of
"folding" the windings is to maintain the length of the hinge in a
more compact space. Thereby, the desired degree of flexibility in
the hinge is maintained in a small space. Another way of describing
the pairs of radial serpentine hinges 704, 706 is to say that the
windings of the hinges have parallel arms. This means that the arms
of the hinges extend in a direction substantially parallel to the
axis of rotation. This is in contrast to the arms of an embodiment
like that depicted in FIG. 5(a) where the arms can be said to be
transverse to the axis of rotation.
Another embodiment 800 is depicted in FIG. 8(a). This embodiment is
also capable of compactly arranging a mirror 807, two sets of
hinges 804 and 806, and frame 808 into as small area as possible.
In the depicted embodiment the pairs of serpentine hinges 804, 806
are circumferentially curved. Such circumferentially curved
serpentine hinges 804, 806 are generally contoured to coincide with
the shape of the outside edge of the mirror 807. Each of the
depicted circumferentially curved serpentine hinges 804, 806 has
four windings comprising a circumferentially curved "quad"
serpentine hinge. As with the other embodiments the hinges can have
any number (n) of windings.
FIG. 8(b) shows a layer 801 of the reflector assembly 800 which
lies just underneath the mirror/frame/hinge structure depicted in
FIG. 8(a). FIG. 8(b) depicts the drive elements 807', 807'', 808',
and 808'. As with the previous embodiments, the drive elements
807', 807'' interact with the mirror 807 to provide deflection
about a first axis. Also, the drive elements 808', 808' interact
with the frame 808 to provide deflection about a second axis.
FIG. 9(a) illustrates another preferred hinge embodiment. The
depicted hinge 900 is a variable spring constant serpentine hinge.
Such a variable spring constant serpentine hinge causes vibrational
damping in the hinge. In some embodiments the implementation of
such damping means is highly desirable. The depicted hinge 900
includes four windings. The hinge 900 begins with the longest arms
on the winding at one end of the hinge 900 and the shortest arms at
the other end of the hinge 900. The arms of each successive winding
are progressively shorter than that of the previous winding. Thus,
winding 922 is shorter than winding 921. In like manner, winding
923 is shorter than winding 922 and winding 924 is shorter than
winding 923. Such variable spring constant serpentine hinges 900
improve the resonant and vibrational behaviour of the optical
elements suspended by the hinges. As with other hinges discussed
herein, the number of windings is variable and determined by the
designer prior to fabrication. The variable spring constant
serpentine hinges 900 can be applied to any of the embodiments
discussed herein. Such hinges have particular utility when applied
to embodiments like that depicted in FIG. 9(b).
FIG. 9(b) depicts a reflector assembly embodiment 901 having pairs
of variable spring constant serpentine hinges 904, 906. As with the
embodiment of FIG. 8(a) the hinges are circumferentially curved. In
addition to being generally contoured to coincide with the shape of
the outside edge of the mirror 907, the circumferentially curved
serpentine hinges 904, 906 are constructed such that they
demonstrate a variable spring constant in the hinges. Each of the
hinges 904, 906 of depicted embodiment includes two windings. As
with all of the other embodiments discussed herein, the hinges can
comprise any number of windings. In the depicted embodiment, the
arms of the windings nearest to the mirror 907 are longer than the
arms of the windings further from the mirror 907. In embodiments
having a greater number of windings in the hinges, the windings are
formed of progressively shorter arm lengths until the desired
resonance and vibration behavior is obtained for the hinge. FIG. 9
shows driving electrodes corresponding to reflector in FIG. 9(b).
Electrodes 907' and 907'' are used to deflect the mirror 907 while
electrodes 908' and 908'' are used to deflect the frame 908.
Another reflector assembly 1000 embodiment is depicted in FIG.
10(a). FIG. 10(a) depicts an embodiment utilizing combinations of
serpentine hinges 1071, 1072 and short straight hinges 1081, 1082.
FIG. 10(a) is a plan view illustrating one embodiment of a
reflector assembly 1000 in accordance with the principles of the
present invention. As with the previous embodiments, the reflector
assembly 1000 typically is incorporated into an array of
reflectors. Each reflector assembly 1000 is fabricated on a
reflector array substrate 1065.
The embodiment 1000 includes a first frame 1010 which connected to
the substrate 1065 by a pair of first serpentine frame hinges 1071
which allows the first frame 1010 to rotate about a first axis
defined by the first serpentine frame hinges 1071. The first frame
1010 is constructed having an inside periphery 1100 and an outside
periphery 1100'. The first serpentine frame hinges 1071 connect the
outside periphery 1100' of the first frame 1010 to the substrate
1065. Positioned inside the first frame 1010 is a second frame
1008. The second frame 1008 includes an inside periphery 1080 and
an outside periphery 1080'. The second frame 1008 is suspended and
supported by a pair of first straight hinges 1081 that allow the
second frame 1008 to rotate about an axis substantially parallel to
the first axis defined by the pair of first serpentine frame hinges
1071. Positioned inside the second frame 1008 is a third frame
1009. The third frame 1009 also includes an inside periphery 1090
and an outside periphery 1090'. The third frame 1009 is suspended
and supported by a pair of second serpentine frame hinges 1072
which connects the outside periphery 1090' of the third frame 1009
to the inside periphery 1080 of the second frame 1008. The pair of
second serpentine frame hinges 1072 allows the third frame 1009 to
rotate about a second axis defined by the pair of second serpentine
frame hinges 1072. The second axis is typically transverse to the
first axis. In a preferred embodiment the second axis is at a
substantially right angle to the first axis. Positioned inside the
third frame 1009 is a mirror 1007. The mirror 1007 includes an
outside periphery 1070. The mirror 1007 is suspended and supported
by a pair of second straight hinges 1082 that allows the mirror
1007 to rotate about an axis substantially parallel to the second
axis defined by the pair of second serpentine frame hinges
1072.
FIG. 10(b) shows a layer of the reflector assembly embodiment 1001
which lies just underneath the mirror/frame/hinge structure
depicted in FIG. 10(a). FIG. 10(b) depicts the multiple drive
elements of the embodiment 1001.
Drive elements 1007' and 1007'' interact with the mirror 1007 to
provide positive and negative deflection about the second axis.
Drive elements 1009' and 1009'' interact with the third frame 1009
to provide added positive and negative deflection about the second
axis.
Drive elements 1008' and 1008'' interact with the second frame 1008
to provide positive and negative deflection about the first axis.
Drive elements 1010' and 1010'' interact with the first frame 1010
to provide added positive and negative deflection about the first
axis.
As previously discussed, the drive elements are shaped and sized
such that they do not interfere with the operation and range of
motion of the hinges 1071, 1072, 1081, 1082. This typically means
that the drive elements 1007', 1007'', 1009', 1009'', 1008',
1008'', 1010', and 1010'' have small cut out regions under the
hinges such that they do not impede hinge operation. Also, as
previously discussed, the drive elements 1007', 1007'', 1009',
1009'', 1008', 1008'', 1010', and 1010'' can be sized such that in
the event of excessive deflection of the movable components (e.g.,
the mirror and frames), no contact is made between the drive
elements and the movable components of the reflector assembly
1000.
The inventors contemplate that the serpentine hinges (e.g., hinges
1071, 1072) shown in the embodiments depicted in FIG. 10(a) and
FIG. 10(b) can easily be replaced by other serpentine hinge
designs. For example, suitable replacements can be the radial
serpentine hinge 704 depicted in FIG. 7(a) or the variable spring
constant serpentine hinge 900 of FIG. 9(a). Such embodiments are to
be taken as illustrative examples rather than limitations. Also,
the hinges of the embodiments depicted FIGS. 10(a) and 10(b) can be
treated with damping agents to improve vibrational and resonance
behavior.
The structures disclosed herein can be can be fabricated out of
silicon based materials using MEMS surface or bulk micromachining
technologies. Examples of such fabrication techniques are discussed
in many standard references. Examples include "Silicon
Micromachining" (1998) by Elwenspoek, M. and Jansen, H. V.; "An
Introduction to Microelectromechanical Systems Engineering" (2000)
Nadim, M.; "Handbook of Microlithography, Micromachining, and
Microfabrication" (1997) Rai-Choudhury, P. Also, a suitable method
of manufacture is discussed in the paper "A Flat High-Frequency
Scanning Micromirror" (2000) Solid-State Sensor & Actuator
Workshop, Hilton Head, S.C., Jun. 4 8, 2000 by Conant, R. A., Nec,
J. T., Lau, K. Y., and Muller, R. S.
Extension of these general fabrication principles from uni-axial
actuators to bi axial actuators, and from structures where both the
reflector and the hinge have the same thickness to devices where
the reflector and hinge thicknesses are different presents a
challenging fabrication problem. This is important because, it is
desirable to have relatively thin hinges, otherwise the hinge
stiffness can be too high requiring large torque to produce the
desired deflection angles, which in turn leads to high driving
signals. However, if the same low thickness is used for the
reflectors, metal coating stress and/or oxide stress can result in
excessive mirror distortion. Therefore, a fabrication process that
permits the decoupling of reflector and hinge thicknesses is
advantageous. In addition, release and separation of these fragile
bi-axial actuators requires special release and separation
techniques.
FIGS. 11(a) 11(m) illustrate a series of cross-section views of a
substrate at selected points in a fabrication process. The process
is depicted with respect to, for example, the device shown in FIG.
6(a) with cross section along line 610. The depicted fabrication
process embodiment can be used to construct bi-axial actuators
having hinge thickness less than reflector thickness.
Alternatively, the hinges can be fabricated having hinge thickness
approximately the same as reflector thickness. Also, the depicted
embodiment is shown having serpentine hinges. The same processes
can be used to fabricate ordinary torsional or bending hinges.
The depicted method embodiment illustrates a fabrication method
using a single layer silicon-on-insulator (SOI) wafer. Referring to
FIG. 11(a), a suitable single layer SOI wafer 1101 can be
fabricated by oxidation and bonding of silicon wafers 1104. These
wafers can be treated using known processes to produce SOI wafer
1101 having a device silicon layer 1102, internal oxide layer 1103,
and silicon wafer layer 1104. A typical thickness of wafer 1104
being on the order of about 300 to 500 um, although wafers having
other thicknesses can be used. The internal oxide layer 1103 is
fabricated on the wafer layer 1104. The oxide layer 1103 can be
fabricated by a variety of processes known to those having ordinary
skill in the art to a thickness of less than 2 um. The device
silicon layer 1102 can then be fabricated on the oxide layer 1103
by grinding, lapping and polishing to a thickness in the range of
about 1 um to 100 um, with 20 to 50 um being preferred. Other
fabrication methods of SOI wafers can also be employed with
particular emphasis on fabrication processes that permit the layer
1103 to be of low stress. Layer 1103 can be fabricated using
materials other than silicon dioxide, such materials include, but
are not limited to silicon nitrides, silicon oxynitrides, aluminum
oxides, and other materials that form good bonding with silicon and
are good etch stops in reactive ion etching of silicon.
In FIG. 11(b) both sides of the SOI wafer are treated to form a top
oxide film layer 1111 and a bottom oxide film layer 1112. The top
oxide film layer 1111 and a bottom oxide film layer 1112 are
typically each formed to a thickness of less than or equal to 3
um.
In FIG. 11(c) a bottom photoresist layer 1113 is formed on the
bottom oxide film layer 1112. The photoresist layer 1113 has
openings defining a bottom pocket 1116 and a separation line 1117.
In FIG. 11(d) a top photoresist layer 1114 is formed on the top
oxide film layer 1111. The top photoresist layer 1114 also has
openings 1115 and 1118 formed therein.
FIG. 11(e) illustrates the top and bottom oxide film layers 1111
and 1112 after oxide material has been removed in a first etching
operation. Material is removed in the openings 1115, 1116, 1117,
1118 in the photoresist layers 1113 and 1114. Typically, this is
accomplished using etching techniques known in the art. In one
embodiment, this etching of the oxide layers 1111 and 1112 is
accomplished using wet etching techniques. As is known to one of
ordinary skill in the art, dry etch techniques can be used.
FIG. 11(f) illustrates the formation of a second top photoresist
layer 11122. The second top photoresist layer 1122 is formed over
remaining top oxide layer 1111 and over portions of the exposed
device silicon layer 1102 in hinge regions 1120, 1121 (region 1115
of FIG. 11(d)). Certain areas 1118 of exposed device silicon layer
1102 are not masked. The second top photoresist layer 1122 is
patterned in the hinge regions 1120, 1121 to permit the formation
of hinges by etching.
FIG. 11(g) shows the effect of a second etching (material removal)
operation. This operation is typically accomplished using etching.
In particular, reactive ion etching (RIE) or other directional
etching techniques are preferred. This etch step defines the
thickness of hinges in regions 1120 and 1121, and also defines the
difference between reflector thickness and hinge thickness. With
reference to FIG. 11(h) the top photoresist layer 1122 is
removed.
FIG. 11(i) illustrates a third etching operation. The top oxide
layer serves as a hard mask over the device silicon layer 1102. The
exposed regions of the device silicon layer 1102 are etched. In
particular, in hinge regions 1120, 1121 (of FIG. 11(g)) and the
exposed areas 1118. Such etching should be accomplished using RIE
or other directional etch techniques. In this way the patterned
hinge areas 1120, 1121 will maintain their pattern and maintain
their differential thickness with respect to reflector thickness.
The internal oxide layer 1103 serves as an etch stop for the third
etch operation.
FIG. 11(j) illustrates a fourth etching (or material removal)
operation. The bottom surface of the SOI wafer 1101 is etched
through openings in the bottom oxide layer 1112. The fourth etch
removes material to form a pocket in region 1116 and to define
separation lines in region 1117. The material can be removed by
etching, preferably using REI or other directional etching
techniques known to those having ordinary skill in the art. Again,
the internal oxide layer 1103 serves as an etch stop for the fourth
etch operation.
FIG. 11(k) illustrates a fifth etching (or material removal)
operation. The fifth etch removes the internal oxide layer 1103 by
backside etching. Etching techniques known to those having ordinary
skill in the art may be used. FIG. 11(l) depicts the forming of a
reflective layer 1129 on one or both sides of the movable optical
element 1128. The reflective layer can be formed using a wide
variety of materials and techniques known to one of ordinary skill
in the art. One process includes forming a metal reflective layer
1129 on at least one of the top and bottom surfaces of the movable
optical element 1128. A suitable metallization material includes,
but is not limited to gold. Adhesion layers, such as chromium,
titanium or tantalum may be employed. A wide variety of deposition
techniques can be used to form the metal reflective layers 1129,
for example, double sided sputtering.
FIG. 11(m) depicts a sixth etching operation used to remove
material in the region 1117 to complete the separation line 1130.
This allows the actuators to be released from the substrates in
arrays of desired size. An earlier etching of these lines would
lead to a premature separation of the wafer into arrays. Once
separated, the separated arrays can then be coupled and aligned
with a mated wafer having formed thereon interconnect circuitry,
driving electronics, and control circuitry. These completed arrays
are hermetically sealed in packages.
The order of the steps can be altered without departing from the
principles of the invention. The use of oxide masks can be
substituted with additional photoresist masks. Also, low-stress
dielectric materials in layer 1103 facilitate release of structures
from the wafer. Also, it is preferable to use low stress materials
for the internal etch stop layers. Such materials include low
stress silicon oxides on the order of about 10 100 MPa. Sputtering
or plasma enhanced chemical vapor deposition processes that provide
very low stress are used rather than thermally deposited oxides.
Because hinges are fabricated from single crystal silicon, creep
and fatigue are minimized and reliability is improved as compared
with devices that use hinges made with polysilicon, metal and metal
alloys in surface micromachining. Rotational comb designs have
leads incorporated on movable electrodes and no bottom electrodes
are required. The interconnections between the top and bottom
wafers are fabricated using, for example, solder reflow or other
techniques.
It should be noted that the optical devices formed on the wafers
are very delicate. Care must be taken in separating the wafer into
its component arrays. One approach for separating the very
sensitive actuators into individual arrays (dies) is performed in
combination of three steps. First, separation lines are defined
lithographically or with shadow masking and dry etched, usually
using standard deep reactive ion etching of silicon. The etch depth
is chosen such that the wafer containing the actuators retains its
rigidity but does not separate into individual dies. In the next
step, deeper cuts are made along separation lines with laser
cutting. The cut depth is controlled by pulse energy, pulse rate,
number of pulses and translational speed of the substrate or laser
beam. It is desirable to use lasers with very short pulse duration
as shorter pulses reduce size and amount of particulate
contamination. In addition, short wavelength lasers are used in
order to provide sufficient absorption of laser energy by the
material desired to be cut. Examples of the appropriate lasers are
tripled or quadrupled neodymium YAG and Ti sapphire. With very
short laser pulses, only gaseous by-products form during cutting
and thus particulate contamination can be eliminated. Photochemical
laser cutting can also be employed. A small thickness of material
is left remaining in the trenches so that particulate and/or
gaseous contamination does not collect on the more critical
surfaces (e.g. optical reflecting surfaces) during laser of the
device. The final step involves cleaving this remaining material
with a small amount of torque applied to separate the arrays. An
alternative separation process can use only dry etching in
combination with cleaving or laser cutting followed by cleaving. A
preferred approach includes all three process steps. Additionally,
these techniques, either individually or in combination, can be
used to effect device separation from both the front and the
backside of the wafer.
The present invention has been particularly shown and described
with respect to certain preferred embodiments and specific features
thereof. However, it should be readily apparent to those of
ordinary skill in the art that various changes and modifications in
form and detail may be made without departing from the spirit and
scope of the invention as set forth in the appended claims. In
particular, it is contemplated by the inventors that the various
hinge types disclosed herein can be interchanged in the various
array embodiments. Also, the reflector array embodiments disclosed
herein can be practiced with optical switch embodiments having one,
two, three, and more reflector arrays. Also, the principles of the
present invention may be practiced with reflectors having other
structures and reflector geometries. Furthermore, the examples
provided herein are intended to be illustrative rather than
limiting. The inventions illustratively disclosed herein can be
practiced without any element which is not specifically disclosed
herein.
* * * * *